[gentoo-user] system uptime

[gentoo-user] system uptime

[Philip Webb]

How long do desktop users typically leave their systems between reboots ?
How long between power off/on's ?

I've long been in the habit of switching everything off while I sleep,
then restarting after I've woken & got going again myself.
However recently, I've run into delays getting my router
(only  1  device attached) to shake hands successfully with my ISP's server,
which have been requiring several power off/on's before it works.
As a result, I've started rebooting only after my weekly system update
-- it means I get to use the new versions of everything --
& not powering off at all ; the monitor + Xscreensaver are off
whenever I'm away from the machine for  >= 1 hr  (approx).

Are there any pro's/con's I sb aware of ?

-- 
========================,,============================================
SUPPORT     ___________//___,   Philip Webb
ELECTRIC   /] [] [] [] [] []|   Cities Centre, University of Toronto
TRANSIT    `-O----------O---'   purslowatchassdotutorontodotca

Re: [gentoo-user] system uptime

[Dale]

Philip Webb wrote:
> How long do desktop users typically leave their systems between reboots ?
> How long between power off/on's ?
>
> I've long been in the habit of switching everything off while I sleep,
> then restarting after I've woken & got going again myself.
> However recently, I've run into delays getting my router
> (only  1  device attached) to shake hands successfully with my ISP's server,
> which have been requiring several power off/on's before it works.
> As a result, I've started rebooting only after my weekly system update
> -- it means I get to use the new versions of everything --
> & not powering off at all ; the monitor + Xscreensaver are off
> whenever I'm away from the machine for  >= 1 hr  (approx).
>
> Are there any pro's/con's I sb aware of ?
>


root@fireball / # uprecords
     #               Uptime | System                                    
Boot up
----------------------------+---------------------------------------------------
     1   193 days, 09:28:37 | Linux 3.5.3-gentoo        Sat Sep 22
07:50:38 2012
     2   116 days, 16:24:24 | Linux 3.16.3-gentoo       Mon Oct 13
20:27:52 2014
     3   111 days, 00:34:49 | Linux 3.18.7-gentoo       Tue Mar 31
18:57:19 2015
     4   101 days, 18:34:17 | Linux 3.5.3-gentoo        Wed Dec 31
18:00:00 1969
     5    72 days, 12:03:16 | Linux 3.9.5-gentoo        Sat Jul 13
19:11:24 2013
     6    69 days, 00:44:23 | Linux 3.11.6-gentoo       Mon Jan  6
03:33:34 2014
     7    66 days, 11:00:52 | Linux 3.9.5-gentoo        Thu Oct 31
15:54:27 2013
     8    51 days, 23:49:06 | Linux 3.13.6-gentoo       Sun Mar 23
15:53:30 2014
     9    46 days, 01:07:54 | Linux 3.16.0-gentoo       Thu Aug 28
15:48:57 2014
    10    36 days, 11:40:14 | Linux 3.14.0-gentoo       Mon May 19
16:05:48 2014
----------------------------+---------------------------------------------------
->  28     6 days, 20:58:28 | Linux 3.18.7-gentoo       Sun Aug 23
02:14:26 2015
----------------------------+---------------------------------------------------
1up in     1 day , 01:40:30 | at                        Mon Aug 31
00:53:23 2015
t10 in    29 days, 14:41:47 | at                        Mon Sep 28
13:54:40 2015
no1 in   186 days, 12:30:10 | at                        Thu Mar  3
10:43:03 2016
    up  1179 days, 07:42:13 | since                     Wed Dec 31
18:00:00 1969
  down  15497 days, 20:30:4 | since                     Wed Dec 31
18:00:00 1969
   %up                7.071 | since                     Wed Dec 31
18:00:00 1969
root@fireball / #


The biggest reason I shutdown, power failure.  I use checkrestart to see
if/when I need to restart something after doing updates.  If for example
I update something in the @system area, then I just logout of the GUI,
go to boot runlevel, run checkrestart again to see if that did it and
then go back to default runlevel.  Sometimes, I have to restart
something by hand instead of rebooting but not to often.  Generally just
going to boot runlevel gets the job done.

One thing about not rebooting a lot, you use cache a lot which can speed
some things up a bit.  I have 16GBs here and most of the time, it is
almost all used.  How much that helps, I dunno but if it didn't help,
they wouldn't have it doing it.  Another good side, run updates while
you sleep. 

The only bad side, more wear on things like fans and some extra dust.  I
try to clean my rig at least twice a year or whenever I notice the temps
a little higher than they should be.  Oh, pulls power all the time which
may not matter much depending on your electricity rates. 

Of course, fixing that connection issue may be a good idea too.  ;-)

Dale

:-)  :-) 

Re: [gentoo-user] system uptime

[covici]

Dale <rdalek1967@gmail.com> wrote:

> Philip Webb wrote:
> > How long do desktop users typically leave their systems between reboots ?
> > How long between power off/on's ?
> >
> > I've long been in the habit of switching everything off while I sleep,
> > then restarting after I've woken & got going again myself.
> > However recently, I've run into delays getting my router
> > (only  1  device attached) to shake hands successfully with my ISP's server,
> > which have been requiring several power off/on's before it works.
> > As a result, I've started rebooting only after my weekly system update
> > -- it means I get to use the new versions of everything --
> > & not powering off at all ; the monitor + Xscreensaver are off
> > whenever I'm away from the machine for  >= 1 hr  (approx).
> >
> > Are there any pro's/con's I sb aware of ?
> >
> 
> 
> root@fireball / # uprecords
>      #               Uptime | System                                    
> Boot up
> ----------------------------+---------------------------------------------------
>      1   193 days, 09:28:37 | Linux 3.5.3-gentoo        Sat Sep 22
> 07:50:38 2012
>      2   116 days, 16:24:24 | Linux 3.16.3-gentoo       Mon Oct 13
> 20:27:52 2014
>      3   111 days, 00:34:49 | Linux 3.18.7-gentoo       Tue Mar 31
> 18:57:19 2015
>      4   101 days, 18:34:17 | Linux 3.5.3-gentoo        Wed Dec 31
> 18:00:00 1969
>      5    72 days, 12:03:16 | Linux 3.9.5-gentoo        Sat Jul 13
> 19:11:24 2013
>      6    69 days, 00:44:23 | Linux 3.11.6-gentoo       Mon Jan  6
> 03:33:34 2014
>      7    66 days, 11:00:52 | Linux 3.9.5-gentoo        Thu Oct 31
> 15:54:27 2013
>      8    51 days, 23:49:06 | Linux 3.13.6-gentoo       Sun Mar 23
> 15:53:30 2014
>      9    46 days, 01:07:54 | Linux 3.16.0-gentoo       Thu Aug 28
> 15:48:57 2014
>     10    36 days, 11:40:14 | Linux 3.14.0-gentoo       Mon May 19
> 16:05:48 2014
> ----------------------------+---------------------------------------------------
> ->  28     6 days, 20:58:28 | Linux 3.18.7-gentoo       Sun Aug 23
> 02:14:26 2015
> ----------------------------+---------------------------------------------------
> 1up in     1 day , 01:40:30 | at                        Mon Aug 31
> 00:53:23 2015
> t10 in    29 days, 14:41:47 | at                        Mon Sep 28
> 13:54:40 2015
> no1 in   186 days, 12:30:10 | at                        Thu Mar  3
> 10:43:03 2016
>     up  1179 days, 07:42:13 | since                     Wed Dec 31
> 18:00:00 1969
>   down  15497 days, 20:30:4 | since                     Wed Dec 31
> 18:00:00 1969
>    %up                7.071 | since                     Wed Dec 31
> 18:00:00 1969
> root@fireball / #
> 
> 
> The biggest reason I shutdown, power failure.  I use checkrestart to see
> if/when I need to restart something after doing updates.  If for example
> I update something in the @system area, then I just logout of the GUI,
> go to boot runlevel, run checkrestart again to see if that did it and
> then go back to default runlevel.  Sometimes, I have to restart
> something by hand instead of rebooting but not to often.  Generally just
> going to boot runlevel gets the job done.
> 
> One thing about not rebooting a lot, you use cache a lot which can speed
> some things up a bit.  I have 16GBs here and most of the time, it is
> almost all used.  How much that helps, I dunno but if it didn't help,
> they wouldn't have it doing it.  Another good side, run updates while
> you sleep. 
> 
> The only bad side, more wear on things like fans and some extra dust.  I
> try to clean my rig at least twice a year or whenever I notice the temps
> a little higher than they should be.  Oh, pulls power all the time which
> may not matter much depending on your electricity rates. 
> 
> Of course, fixing that connection issue may be a good idea too.  ;-)

hmmm, if you go to boot run level what is the difference between that
and rebooting?  After a major update there are so many things to restart
that I usually give up and reboot the system, is actually quicker.


-- 
Your life is like a penny.  You're going to lose it.  The question is:
How do
you spend it?

         John Covici
         covici@ccs.covici.com

Re: [gentoo-user] system uptime

[Dale]

covici@ccs.covici.com wrote:
> Dale <rdalek1967@gmail.com> wrote:
>
>
> The biggest reason I shutdown, power failure.  I use checkrestart to see
> if/when I need to restart something after doing updates.  If for example
> I update something in the @system area, then I just logout of the GUI,
> go to boot runlevel, run checkrestart again to see if that did it and
> then go back to default runlevel.  Sometimes, I have to restart
> something by hand instead of rebooting but not to often.  Generally just
> going to boot runlevel gets the job done.
>
> One thing about not rebooting a lot, you use cache a lot which can speed
> some things up a bit.  I have 16GBs here and most of the time, it is
> almost all used.  How much that helps, I dunno but if it didn't help,
> they wouldn't have it doing it.  Another good side, run updates while
> you sleep. 
>
> The only bad side, more wear on things like fans and some extra dust.  I
> try to clean my rig at least twice a year or whenever I notice the temps
> a little higher than they should be.  Oh, pulls power all the time which
> may not matter much depending on your electricity rates. 
>
> Of course, fixing that connection issue may be a good idea too.  ;-)

> hmmm, if you go to boot run level what is the difference between that
> and rebooting?  After a major update there are so many things to restart
> that I usually give up and reboot the system, is actually quicker.
>
>

Hmmmm, this quoting thing didn't work right again. 

For me, it is faster.  Also, rebooting can uncover a problem that I
might not know about.  I've had a few times where I couldn't reboot for
some unknown reason.  Plus, all the common stuff remains in cache.  Most
of the time tho, just logging out of a GUI, KDE for me, is enough. 
Using checkrestart should tell me exactly what needs to be restarted and
most of the time how.  About the only thing I have to restart manually,
udev.  It's one thing that has a regular update that doesn't restart
since it is already started before getting to the boot runlevel. 

To each his own tho.  All of us has our own way of doing things of this
nature and for varying reasons.  Some shutdown because electricity is
expensive.  For some, that doesn't matter.  Some do it to just reduce
noise from the fans etc.  One reason I leave mine on all the time is
that I almost always have mine doing something.  I have tons of TV shows
and such on here.  If I'm not doing something myself, I have it doing
something. 

Dale

:-)  :-) 

Re: [gentoo-user] system uptime

[Mick]

[-- Attachment #1: Type: Text/Plain, Size: 2800 bytes --]

On Sunday 30 Aug 2015 08:54:16 Dale wrote:
> covici@ccs.covici.com wrote:
> > Dale <rdalek1967@gmail.com> wrote:
> > 
> > 
> > The biggest reason I shutdown, power failure.  I use checkrestart to see
> > if/when I need to restart something after doing updates.  If for example
> > I update something in the @system area, then I just logout of the GUI,
> > go to boot runlevel, run checkrestart again to see if that did it and
> > then go back to default runlevel.  Sometimes, I have to restart
> > something by hand instead of rebooting but not to often.  Generally just
> > going to boot runlevel gets the job done.
> > 
> > One thing about not rebooting a lot, you use cache a lot which can speed
> > some things up a bit.  I have 16GBs here and most of the time, it is
> > almost all used.  How much that helps, I dunno but if it didn't help,
> > they wouldn't have it doing it.  Another good side, run updates while
> > you sleep.
> > 
> > The only bad side, more wear on things like fans and some extra dust.  I
> > try to clean my rig at least twice a year or whenever I notice the temps
> > a little higher than they should be.  Oh, pulls power all the time which
> > may not matter much depending on your electricity rates.
> > 
> > Of course, fixing that connection issue may be a good idea too.  ;-)
> > 
> > hmmm, if you go to boot run level what is the difference between that
> > and rebooting?  After a major update there are so many things to restart
> > that I usually give up and reboot the system, is actually quicker.
> 
> Hmmmm, this quoting thing didn't work right again.
> 
> For me, it is faster.  Also, rebooting can uncover a problem that I
> might not know about.  I've had a few times where I couldn't reboot for
> some unknown reason.  Plus, all the common stuff remains in cache.  Most
> of the time tho, just logging out of a GUI, KDE for me, is enough.
> Using checkrestart should tell me exactly what needs to be restarted and
> most of the time how.  About the only thing I have to restart manually,
> udev.  It's one thing that has a regular update that doesn't restart
> since it is already started before getting to the boot runlevel.
> 
> To each his own tho.  All of us has our own way of doing things of this
> nature and for varying reasons.  Some shutdown because electricity is
> expensive.  For some, that doesn't matter.  Some do it to just reduce
> noise from the fans etc.  One reason I leave mine on all the time is
> that I almost always have mine doing something.  I have tons of TV shows
> and such on here.  If I'm not doing something myself, I have it doing
> something.
> 
> Dale
> 
> :-)  :-)

What do you do if you install a new kernel?  You have to reboot then, yes?

-- 
Regards,
Mick

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 473 bytes --]

Re: [gentoo-user] system uptime

[Dale]

Mick wrote:
> On Sunday 30 Aug 2015 08:54:16 Dale wrote:
>> covici@ccs.covici.com wrote:
>>> Dale <rdalek1967@gmail.com> wrote:
>>>
>>>
>>> The biggest reason I shutdown, power failure.  I use checkrestart to see
>>> if/when I need to restart something after doing updates.  If for example
>>> I update something in the @system area, then I just logout of the GUI,
>>> go to boot runlevel, run checkrestart again to see if that did it and
>>> then go back to default runlevel.  Sometimes, I have to restart
>>> something by hand instead of rebooting but not to often.  Generally just
>>> going to boot runlevel gets the job done.
>>>
>>> One thing about not rebooting a lot, you use cache a lot which can speed
>>> some things up a bit.  I have 16GBs here and most of the time, it is
>>> almost all used.  How much that helps, I dunno but if it didn't help,
>>> they wouldn't have it doing it.  Another good side, run updates while
>>> you sleep.
>>>
>>> The only bad side, more wear on things like fans and some extra dust.  I
>>> try to clean my rig at least twice a year or whenever I notice the temps
>>> a little higher than they should be.  Oh, pulls power all the time which
>>> may not matter much depending on your electricity rates.
>>>
>>> Of course, fixing that connection issue may be a good idea too.  ;-)
>>>
>>> hmmm, if you go to boot run level what is the difference between that
>>> and rebooting?  After a major update there are so many things to restart
>>> that I usually give up and reboot the system, is actually quicker.
>> Hmmmm, this quoting thing didn't work right again.
>>
>> For me, it is faster.  Also, rebooting can uncover a problem that I
>> might not know about.  I've had a few times where I couldn't reboot for
>> some unknown reason.  Plus, all the common stuff remains in cache.  Most
>> of the time tho, just logging out of a GUI, KDE for me, is enough.
>> Using checkrestart should tell me exactly what needs to be restarted and
>> most of the time how.  About the only thing I have to restart manually,
>> udev.  It's one thing that has a regular update that doesn't restart
>> since it is already started before getting to the boot runlevel.
>>
>> To each his own tho.  All of us has our own way of doing things of this
>> nature and for varying reasons.  Some shutdown because electricity is
>> expensive.  For some, that doesn't matter.  Some do it to just reduce
>> noise from the fans etc.  One reason I leave mine on all the time is
>> that I almost always have mine doing something.  I have tons of TV shows
>> and such on here.  If I'm not doing something myself, I have it doing
>> something.
>>
>> Dale
>>
>> :-)  :-)
> What do you do if you install a new kernel?  You have to reboot then, yes?
>


Of course.  Don't you?  I just don't have a huge need to update the
kernel that often.  I'm not running some server that has to worry about
getting hacked 10,000 times a day.  I just update it when I can. 

I might add, I'm stuck on the current kernel because NONE of the newer
ones will boot.  There's another thread on that where someone else has
the issue.  So, until that is fixed and I CAN update, no need worrying
about a new kernel needing to be loaded.  That just leaves me with power
failures and such. 

Dale

:-)  :-) 

Re: [gentoo-user] system uptime

[Walter Dnes]

On Sun, Aug 30, 2015 at 12:04:43AM -0400, Philip Webb wrote
> How long do desktop users typically leave their systems between reboots ?
> How long between power off/on's ?
> 
> I've long been in the habit of switching everything off while I sleep,
> then restarting after I've woken & got going again myself.  However
> recently, I've run into delays getting my router (only  1  device
> attached) to shake hands successfully with my ISP's server, which have
> been requiring several power off/on's before it works.

  I'm on Teksavvy, and I run into that on occasion.

> As a result, I've started rebooting only after my weekly system update
> -- it means I get to use the new versions of everything -- & not
> powering off at all ; the monitor + Xscreensaver are off whenever I'm
> away from the machine for  >= 1 hr  (approx).
> 
> Are there any pro's/con's I sb aware of ?

[d531][waltdnes][~] uptime
 02:14:01 up 39 days,  5:31, 22 users,  load average: 0.16, 0.22, 0.48

  No, my machine has not been on for over 900 consecutive hours.  It's
that long since my most recent "full boot".  sys-power/hibernate-script
in suspend-to-disc mode totally shuts down the machine.  It has to read
the BIOS on start-up, but it restores all workspaces, and program state
with multiple browsers/spreadsheets/etc open, from swap.  I have
multiple browser profiles, allowing me to dedicate separate instances to
each forum.  Plus I have ongong personal projects that have spreadsheets
or vim open.  It's an absolute pain to re-open all the
browsers/spreadsheets/etc in each workspace when I do a "real reboot"
for a new kernel.

  I currently have the display, speakers, modem, router, etc plugged
into power bars that are plugged into a "slave jack" on my UPS.  The
desktop PC is plugged into the "master jack".  When the "master" is
drawing power, the "slave" jack provides power to the power bars.  When
I hibernate the PC, and it powers down, the "slave" jack cuts off power
to the power bars.  So shutting down or hibernating my PC shuts down
display, speakers, modem, router, etc.  Turning the PC back on powers
them up again.  If I had your problem, I would move my router/modem to a
"filtered" plug on the UPS.  So hibernation would shut down everything
except the router/modem.

-- 
Walter Dnes <waltdnes@waltdnes.org>
I don't run "desktop environments"; I run useful applications

Re: [gentoo-user] system uptime

[Alan McKinnon]

On 30/08/2015 06:04, Philip Webb wrote:
> How long do desktop users typically leave their systems between reboots ?
> How long between power off/on's ?
> 
> I've long been in the habit of switching everything off while I sleep,
> then restarting after I've woken & got going again myself.
> However recently, I've run into delays getting my router
> (only  1  device attached) to shake hands successfully with my ISP's server,
> which have been requiring several power off/on's before it works.
> As a result, I've started rebooting only after my weekly system update
> -- it means I get to use the new versions of everything --
> & not powering off at all ; the monitor + Xscreensaver are off
> whenever I'm away from the machine for  >= 1 hr  (approx).
> 
> Are there any pro's/con's I sb aware of ?
> 


Mine depends. I typically do a deliberate reboot only when wanting a new
kernel running. Sometimes it's a few days, often up to a month or more.

Overnight I usually suspend the machine. This works for me because I
usually have a standard array of apps spread over 6 virtual desktops and
I forget how everything was set up (advancing age...)

It's very seldom that an emerge world needs a reboot so that isn't
included in the above. checkrestart and sometimes a log out/log in takes
care of getting updated software to run.

This is Gentoo not windows, so I see no benefit to frequent daily
reboots. Monthly or so, or when new kernels are advised, suits my needs.

An aside: I've also proudly played the uptime game, and had remote DNS
servers with 1600 days uptime. It looks impressive, but all it really
proves is I'm a short-sighted idiot who doesn't do kernel updates :-)

So I don't do that extreme anymore.

-- 
Alan McKinnon
alan.mckinnon@gmail.com

Re: [gentoo-user] system uptime

[Rich Freeman]

On Sun, Aug 30, 2015 at 5:41 AM, Alan McKinnon <alan.mckinnon@gmail.com> wrote:
> On 30/08/2015 06:04, Philip Webb wrote:
>> How long do desktop users typically leave their systems between reboots ?
>> How long between power off/on's ?
>>
>
> Mine depends. I typically do a deliberate reboot only when wanting a new
> kernel running. Sometimes it's a few days, often up to a month or more.
>

++

Typically I reboot every week or two.  I try to keep up with stable
kernel releases on the latest longterm branch (currently 3.18.20).
I'm currently at 19 days, which is a bit on the high side for me, but
it seems like 3.18 hasn't had as many updates as some of the other
stable series.

If you keep a closer eye on security issues and care to track which
kernel fixes you do or don't accept, or want to mess with kernel live
patching, then you could go longer.  I also like to reboot at some
frequency just so that if for whatever reason it doesn't reboot I only
have a few week's worth of system updates to look at to figure out
what changed.  I have Gentoo hosts that I don't stay on top of as
often and when 5 things break at once I'm playing guessing games
(those hosts are all easy to snapshot).

-- 
Rich

Re: [gentoo-user] system uptime

[Michel Catudal]

Le 2015-08-30 00:04, Philip Webb a écrit :
> How long do desktop users typically leave their systems between reboots ?
> How long between power off/on's ?
>
> I've long been in the habit of switching everything off while I sleep,
> then restarting after I've woken & got going again myself.
> However recently, I've run into delays getting my router
> (only  1  device attached) to shake hands successfully with my ISP's server,
> which have been requiring several power off/on's before it works.
> As a result, I've started rebooting only after my weekly system update
> -- it means I get to use the new versions of everything --
> & not powering off at all ; the monitor + Xscreensaver are off
> whenever I'm away from the machine for  >= 1 hr  (approx).
>
> Are there any pro's/con's I sb aware of ?
>
For reboots many users may choose to reboot when they do changes, perhaps habit from windows or OS/2. It is usually not necessary unless you change your kernel or bootloader.

As for shutdowns there are several arguments for and against. What often kills electronic is the shock between hot and cold so there is an argument about keeping the system on.
Whether it is always safe to keep the computer on all the time remains to be proven. My son always leaves his computer on and I had to recently replace it, the mother board was gone. Mine which was purchased around the same time has had no issues, I 
shutdown every night unless I need to do some updates. An argument against it would be wasting energy. Computers are cheap, so are hard disk. Unless you run a server that has to be on all the time there is no logic in keeping the computer on unless you can 
get it to go sleep.

-- 
For Linux Software visit
http://home.comcast.net/~mcatudal
http://sourceforge.net/projects/suzielinux/

Re: [gentoo-user] system uptime

[Daniel Frey]

On 08/30/2015 06:24 AM, Michel Catudal wrote:
> 
> As for shutdowns there are several arguments for and against. What often
> kills electronic is the shock between hot and cold so there is an
> argument about keeping the system on.
> Whether it is always safe to keep the computer on all the time remains
> to be proven. 

Recently I've had to help someone migrate off of a failed computer. This
computer was old (I had to find an IDE adapter to recover some files)
from late 90s/early 00s.

Some time ago I told him to have it running all the time, mostly because
of age. So he kept it running nonstop and literally a week or two ago
shut it down as he was getting new flooring installed. He called me
after hooking it back up again as it wouldn't start. I went over to
check and the motherboard finally failed. He hadn't powered it off in
4-5 years.

For myself I use a smart power bar and suspend my PC when not in use.
This caused me all sorts of grief with systemd hanging on shutdown after
a suspend, ultimately causing my RAID array to be rebuilt on every
reboot/shutdown and so I've finally abandoned it and am running openrc
again.

The only thing about using suspend is that if the PC is in a sleep state
it won't wake up and shut down when the power goes out. This just
happened to me yesterday (big wind storm here.)

Dan

Re: [gentoo-user] system uptime

[Alan McKinnon]

On 30/08/2015 17:24, Daniel Frey wrote:
> On 08/30/2015 06:24 AM, Michel Catudal wrote:
>>
>> As for shutdowns there are several arguments for and against. What often
>> kills electronic is the shock between hot and cold so there is an
>> argument about keeping the system on.
>> Whether it is always safe to keep the computer on all the time remains
>> to be proven. 
> 
> Recently I've had to help someone migrate off of a failed computer. This
> computer was old (I had to find an IDE adapter to recover some files)
> from late 90s/early 00s.
> 
> Some time ago I told him to have it running all the time, mostly because
> of age. So he kept it running nonstop and literally a week or two ago
> shut it down as he was getting new flooring installed. He called me
> after hooking it back up again as it wouldn't start. I went over to
> check and the motherboard finally failed. He hadn't powered it off in
> 4-5 years.
> 
> For myself I use a smart power bar and suspend my PC when not in use.
> This caused me all sorts of grief with systemd hanging on shutdown after
> a suspend, ultimately causing my RAID array to be rebuilt on every
> reboot/shutdown and so I've finally abandoned it and am running openrc
> again.
> 
> The only thing about using suspend is that if the PC is in a sleep state
> it won't wake up and shut down when the power goes out. This just
> happened to me yesterday (big wind storm here.)



One of the reasons sysadmins have old servers out there that still have
huge uptimes, is that we dare not switch them off. We don't know if the
drives will spin up again from cold!

Technically, we should do a power down test every 6 months or so, but
that turns out not to be a yes/no test in real life; it's a yes/destroy
test and no-one wants to make a decision either way. So we all sit in
limbo and wait for some exterior event to decide for us (like black-outs)

Sad, init?


-- 
Alan McKinnon
alan.mckinnon@gmail.com

Re: [gentoo-user] system uptime

[Rich Freeman]

On Sun, Aug 30, 2015 at 11:54 AM, Alan McKinnon <alan.mckinnon@gmail.com> wrote:
> Technically, we should do a power down test every 6 months or so, but
> that turns out not to be a yes/no test in real life; it's a yes/destroy
> test and no-one wants to make a decision either way. So we all sit in
> limbo and wait for some exterior event to decide for us (like black-outs)
>

Half the time these are ancient services that have long been replaced
but nobody can bring themselves to make the call to get rid of the old
servers.  Maybe there were 10M records in the database and 9.998M of
them were migrated to a new database, but due to some issue the rest
couldn't be, so the old server stays up just in case anybody ever
needs the old data, and so on.

Typically these would just stick around until finally some hardware
component fails, and then it gets written off.

Sadly, this course of forcing hands seems to be going away.  At work
somebody tried to hand me an ancient system to look after in my spare
time.  Apparently they just finished virtualizing it.  Go figure -
they have VAX VMs available for Linux these days.  The problem is that
KT and maintaining documentation and not being the person who gets the
finger pointed at when something goes wrong costs the company time and
money, and in this case for almost zero value.

Usually the problems with technology aren't technical in nature...

-- 
Rich

Re: [gentoo-user] system uptime

[Alan Mackenzie]

Hello, Philip.

On Sun, Aug 30, 2015 at 12:04:43AM -0400, Philip Webb wrote:
> How long do desktop users typically leave their systems between reboots ?
> How long between power off/on's ?

I switch my PC on in the morning when I want to start using it.  I
switch it off before going to bed, or going out shopping, or things like
that.  I don't like wasting electricity.

> I've long been in the habit of switching everything off while I sleep,
> then restarting after I've woken & got going again myself.
> However recently, I've run into delays getting my router
> (only  1  device attached) to shake hands successfully with my ISP's server,
> which have been requiring several power off/on's before it works.

Unfortunately, I need to leave my router permanently powered up, which I
resent.  German telephone companies have been offering only VOIP
telephone connections for some while, now, which basically means that
instead of the companies converting IP packets into a telephone signal,
every subscriber's got to do it himself.  So where people could
previously simply buy a telephone handset and plug it into the wall,
they've now potentially got to spend extra on a router and somehow
manage to configure that router.  And of course, the router contuously
wastes electricity, waiting for that occasional incoming call, whereas
previously the handset was only powered up, from the exchange, when a
call was in progress.  Progress this is not.

> As a result, I've started rebooting only after my weekly system update
> -- it means I get to use the new versions of everything --
> & not powering off at all ; the monitor + Xscreensaver are off
> whenever I'm away from the machine for  >= 1 hr  (approx).

> Are there any pro's/con's I sb aware of ?

> -- 
> ========================,,============================================
> SUPPORT     ___________//___,   Philip Webb
> ELECTRIC   /] [] [] [] [] []|   Cities Centre, University of Toronto
> TRANSIT    `-O----------O---'   purslowatchassdotutorontodotca

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: [gentoo-user] system uptime

[Rich Freeman]

On Sun, Aug 30, 2015 at 9:26 AM, Alan Mackenzie <acm@muc.de> wrote:
>
> On Sun, Aug 30, 2015 at 12:04:43AM -0400, Philip Webb wrote:
>> How long do desktop users typically leave their systems between reboots ?
>> How long between power off/on's ?
>
> I switch my PC on in the morning when I want to start using it.  I
> switch it off before going to bed, or going out shopping, or things like
> that.  I don't like wasting electricity.
>

Most discussions I've read on this matter tend to end up around here,
both for economy and wear.  Turn on the computer when you are going to
need it, and turn it off at the end of the day, or similar.  It
probably isn't worth powering it off for lunch (though I do put it to
sleep).  For my server it runs 24x7.  I'm always amused when I look at
the SMART stats on failed drives.  They usually have thousands of
hours of power-on time, and maybe a dozen spin-ups.  Then again, that
one batch of Seagate 1TB drives seemed to die barely broken in
(suffice it to say I've gotten very proficient at their RMA process,
which is decent).

-- 
Rich

Re: [gentoo-user] system uptime

[Mick]

[-- Attachment #1: Type: Text/Plain, Size: 2696 bytes --]

On Sunday 30 Aug 2015 14:26:36 Alan Mackenzie wrote:
> Hello, Philip.
> 
> On Sun, Aug 30, 2015 at 12:04:43AM -0400, Philip Webb wrote:
> > How long do desktop users typically leave their systems between reboots ?
> > How long between power off/on's ?
> 
> I switch my PC on in the morning when I want to start using it.  I
> switch it off before going to bed, or going out shopping, or things like
> that.  I don't like wasting electricity.

Same here.  Unless I am somewhere near the desk the laptop is on sleep.  
Overnight it is shut down.


> > I've long been in the habit of switching everything off while I sleep,
> > then restarting after I've woken & got going again myself.
> > However recently, I've run into delays getting my router
> > (only  1  device attached) to shake hands successfully with my ISP's
> > server, which have been requiring several power off/on's before it
> > works.

I think that it may be better troubleshooting your ISP PPPoE negotiation (or 
whatever protocol they are using) than changing your usage habits.  Well, I 
would be doing this anyway, out of curiosity.  :-)


> Unfortunately, I need to leave my router permanently powered up, which I
> resent.  German telephone companies have been offering only VOIP
> telephone connections for some while, now, which basically means that
> instead of the companies converting IP packets into a telephone signal,
> every subscriber's got to do it himself.  So where people could
> previously simply buy a telephone handset and plug it into the wall,
> they've now potentially got to spend extra on a router and somehow
> manage to configure that router.  And of course, the router contuously
> wastes electricity, waiting for that occasional incoming call, whereas
> previously the handset was only powered up, from the exchange, when a
> call was in progress.  Progress this is not.

I Power cuts can cause re-syncs and a lower sync rate for my connection, so I 
leave the router powered up 24-7 and connected to a UPS.  The waste of 
electricity is tiny.


> > As a result, I've started rebooting only after my weekly system update
> > -- it means I get to use the new versions of everything --
> > & not powering off at all ; the monitor + Xscreensaver are off
> > whenever I'm away from the machine for  >= 1 hr  (approx).
> > 
> > Are there any pro's/con's I sb aware of ?

As others have mentioned consider hibernation for overnight purposes (S4) and 
sleep (S3) for when you are away from your desk for longer periods of time. 
However, I would be intrigued as to what might be wrong with the ISP network 
authentication.  ;-)

-- 
Regards,
Mick

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 473 bytes --]

Re: [gentoo-user] system uptime

[Volker Armin Hemmann]

Am 30.08.2015 um 15:26 schrieb Alan Mackenzie:
> Hello, Philip.
>
> On Sun, Aug 30, 2015 at 12:04:43AM -0400, Philip Webb wrote:
>> How long do desktop users typically leave their systems between reboots ?
>> How long between power off/on's ?
> I switch my PC on in the morning when I want to start using it.  I
> switch it off before going to bed, or going out shopping, or things like
> that.  I don't like wasting electricity.
>
>> I've long been in the habit of switching everything off while I sleep,
>> then restarting after I've woken & got going again myself.
>> However recently, I've run into delays getting my router
>> (only  1  device attached) to shake hands successfully with my ISP's server,
>> which have been requiring several power off/on's before it works.
> Unfortunately, I need to leave my router permanently powered up, which I
> resent. 
yeah, a fritzbox needs so much power....

Re: [gentoo-user] system uptime

[Alec Ten Harmsel]

On Sun, Aug 30, 2015 at 12:04:43AM -0400, Philip Webb wrote:
> How long do desktop users typically leave their systems between reboots ?
> How long between power off/on's ?

I generally power off whenever I won't be using my desktop for more than
7-8 hours, since it has a big CPU, lots of RAM, decent GPU, etc. and
hence draws a lot of idle power. So usually the most uptime I get on it
is around 2 or 3 days.

Alec

Re: [gentoo-user] system uptime

[Volker Armin Hemmann]

Am 30.08.2015 um 06:04 schrieb Philip Webb:
> How long do desktop users typically leave their systems between reboots ?
> How long between power off/on's ?
>
> I've long been in the habit of switching everything off while I sleep,
> then restarting after I've woken & got going again myself.
> However recently, I've run into delays getting my router
> (only  1  device attached) to shake hands successfully with my ISP's server,
> which have been requiring several power off/on's before it works.
> As a result, I've started rebooting only after my weekly system update
> -- it means I get to use the new versions of everything --
> & not powering off at all ; the monitor + Xscreensaver are off
> whenever I'm away from the machine for  >= 1 hr  (approx).
>
> Are there any pro's/con's I sb aware of ?
>

suspend to ram.

Only reboot when there is a kernel update I actually install.

Re: [gentoo-user] system uptime

[Peter Humphrey]

On Sunday 30 August 2015 00:04:43 Philip Webb wrote:
> How long do desktop users typically leave their systems between reboots ?
> How long between power off/on's ?
> 
> I've long been in the habit of switching everything off while I sleep,
> then restarting after I've woken & got going again myself.
> However recently, I've run into delays getting my router
> (only  1  device attached) to shake hands successfully with my ISP's server,
> which have been requiring several power off/on's before it works.
> As a result, I've started rebooting only after my weekly system update
> -- it means I get to use the new versions of everything --
> & not powering off at all ; the monitor + Xscreensaver are off
> whenever I'm away from the machine for  >= 1 hr  (approx).
> 
> Are there any pro's/con's I sb aware of ?

No-one has yet mentioned taking backups. I'm still using a brute-force 
approach, in which I shut down each of my two machines once a week to make a 
backup to external disk. Otherwise they're on 24 hours a day running BOINC 
projects. On the desktop PC kmail makes a daily archive of messages, and once 
a day a cron job copies my user directory to /home/<me>.bu/ .

I know it burns energy but I'm prepared to make my small contribution to what 
I think is a good cause.

-- 
Rgds
Peter

Re: [gentoo-user] system uptime

[Alan McKinnon]

On 30/08/2015 17:56, Peter Humphrey wrote:
> On Sunday 30 August 2015 00:04:43 Philip Webb wrote:
>> How long do desktop users typically leave their systems between reboots ?
>> How long between power off/on's ?
>>
>> I've long been in the habit of switching everything off while I sleep,
>> then restarting after I've woken & got going again myself.
>> However recently, I've run into delays getting my router
>> (only  1  device attached) to shake hands successfully with my ISP's server,
>> which have been requiring several power off/on's before it works.
>> As a result, I've started rebooting only after my weekly system update
>> -- it means I get to use the new versions of everything --
>> & not powering off at all ; the monitor + Xscreensaver are off
>> whenever I'm away from the machine for  >= 1 hr  (approx).
>>
>> Are there any pro's/con's I sb aware of ?
> 
> No-one has yet mentioned taking backups. I'm still using a brute-force 
> approach, in which I shut down each of my two machines once a week to make a 
> backup to external disk. Otherwise they're on 24 hours a day running BOINC 
> projects. On the desktop PC kmail makes a daily archive of messages, and once 
> a day a cron job copies my user directory to /home/<me>.bu/ .
> 
> I know it burns energy but I'm prepared to make my small contribution to what 
> I think is a good cause.
> 


A desktop or laptop will typically draw far less power than a single 60W
incandescent bulb. I bet you have quite a lot of those. Even if not, the
CFLs you'll have to give you light at night still draw much much more
than a computer.

If saving energy is your personal driver, then you should be looking at
water heaters, central heaters, aircon and stove as the main culprits.
Everything else, whilst measurable, is a small drop in the bucket and
probably not worth worrying about.

Assuming of course that your computer is a desktop/laptop, and not a 42U
cabinet jam packed full of Dell 2950s

-- 
Alan McKinnon
alan.mckinnon@gmail.com

Re: [gentoo-user] system uptime

[Dale]

Alan McKinnon wrote:
> On 30/08/2015 17:56, Peter Humphrey wrote:
>> On Sunday 30 August 2015 00:04:43 Philip Webb wrote:
>>> How long do desktop users typically leave their systems between reboots ?
>>> How long between power off/on's ?
>>>
>>> I've long been in the habit of switching everything off while I sleep,
>>> then restarting after I've woken & got going again myself.
>>> However recently, I've run into delays getting my router
>>> (only  1  device attached) to shake hands successfully with my ISP's server,
>>> which have been requiring several power off/on's before it works.
>>> As a result, I've started rebooting only after my weekly system update
>>> -- it means I get to use the new versions of everything --
>>> & not powering off at all ; the monitor + Xscreensaver are off
>>> whenever I'm away from the machine for  >= 1 hr  (approx).
>>>
>>> Are there any pro's/con's I sb aware of ?
>> No-one has yet mentioned taking backups. I'm still using a brute-force 
>> approach, in which I shut down each of my two machines once a week to make a 
>> backup to external disk. Otherwise they're on 24 hours a day running BOINC 
>> projects. On the desktop PC kmail makes a daily archive of messages, and once 
>> a day a cron job copies my user directory to /home/<me>.bu/ .
>>
>> I know it burns energy but I'm prepared to make my small contribution to what 
>> I think is a good cause.
>>
>
> A desktop or laptop will typically draw far less power than a single 60W
> incandescent bulb. I bet you have quite a lot of those. Even if not, the
> CFLs you'll have to give you light at night still draw much much more
> than a computer.
>
> If saving energy is your personal driver, then you should be looking at
> water heaters, central heaters, aircon and stove as the main culprits.
> Everything else, whilst measurable, is a small drop in the bucket and
> probably not worth worrying about.
>
> Assuming of course that your computer is a desktop/laptop, and not a 42U
> cabinet jam packed full of Dell 2950s
>


Don't forget the clothes dryer to, if you have one.  Mine is electric
and it pulls as much as my water heater does.  I just don't use it as
much is all. 

Dale

:-)  :-) 

Re: [gentoo-user] system uptime

[Alan McKinnon]

On 30/08/2015 19:00, Dale wrote:
> Alan McKinnon wrote:
>> On 30/08/2015 17:56, Peter Humphrey wrote:
>>> On Sunday 30 August 2015 00:04:43 Philip Webb wrote:
>>>> How long do desktop users typically leave their systems between reboots ?
>>>> How long between power off/on's ?
>>>>
>>>> I've long been in the habit of switching everything off while I sleep,
>>>> then restarting after I've woken & got going again myself.
>>>> However recently, I've run into delays getting my router
>>>> (only  1  device attached) to shake hands successfully with my ISP's server,
>>>> which have been requiring several power off/on's before it works.
>>>> As a result, I've started rebooting only after my weekly system update
>>>> -- it means I get to use the new versions of everything --
>>>> & not powering off at all ; the monitor + Xscreensaver are off
>>>> whenever I'm away from the machine for  >= 1 hr  (approx).
>>>>
>>>> Are there any pro's/con's I sb aware of ?
>>> No-one has yet mentioned taking backups. I'm still using a brute-force 
>>> approach, in which I shut down each of my two machines once a week to make a 
>>> backup to external disk. Otherwise they're on 24 hours a day running BOINC 
>>> projects. On the desktop PC kmail makes a daily archive of messages, and once 
>>> a day a cron job copies my user directory to /home/<me>.bu/ .
>>>
>>> I know it burns energy but I'm prepared to make my small contribution to what 
>>> I think is a good cause.
>>>
>>
>> A desktop or laptop will typically draw far less power than a single 60W
>> incandescent bulb. I bet you have quite a lot of those. Even if not, the
>> CFLs you'll have to give you light at night still draw much much more
>> than a computer.
>>
>> If saving energy is your personal driver, then you should be looking at
>> water heaters, central heaters, aircon and stove as the main culprits.
>> Everything else, whilst measurable, is a small drop in the bucket and
>> probably not worth worrying about.
>>
>> Assuming of course that your computer is a desktop/laptop, and not a 42U
>> cabinet jam packed full of Dell 2950s
>>
> 
> 
> Don't forget the clothes dryer to, if you have one.  Mine is electric
> and it pulls as much as my water heater does.  I just don't use it as
> much is all. 

I forgot about that :-)

Add in almost all laundry appliances and kitchen power tools too...


-- 
Alan McKinnon
alan.mckinnon@gmail.com

Re: [gentoo-user] system uptime

[Mick]

[-- Attachment #1: Type: Text/Plain, Size: 2938 bytes --]

On Sunday 30 Aug 2015 18:05:13 Alan McKinnon wrote:
> On 30/08/2015 19:00, Dale wrote:
> > Alan McKinnon wrote:
> >> On 30/08/2015 17:56, Peter Humphrey wrote:
> >>> On Sunday 30 August 2015 00:04:43 Philip Webb wrote:
> >>>> How long do desktop users typically leave their systems between
> >>>> reboots ? How long between power off/on's ?
> >>>> 
> >>>> I've long been in the habit of switching everything off while I sleep,
> >>>> then restarting after I've woken & got going again myself.
> >>>> However recently, I've run into delays getting my router
> >>>> (only  1  device attached) to shake hands successfully with my ISP's
> >>>> server, which have been requiring several power off/on's before it
> >>>> works. As a result, I've started rebooting only after my weekly
> >>>> system update -- it means I get to use the new versions of everything
> >>>> --
> >>>> & not powering off at all ; the monitor + Xscreensaver are off
> >>>> whenever I'm away from the machine for  >= 1 hr  (approx).
> >>>> 
> >>>> Are there any pro's/con's I sb aware of ?
> >>> 
> >>> No-one has yet mentioned taking backups. I'm still using a brute-force
> >>> approach, in which I shut down each of my two machines once a week to
> >>> make a backup to external disk. Otherwise they're on 24 hours a day
> >>> running BOINC projects. On the desktop PC kmail makes a daily archive
> >>> of messages, and once a day a cron job copies my user directory to
> >>> /home/<me>.bu/ .
> >>> 
> >>> I know it burns energy but I'm prepared to make my small contribution
> >>> to what I think is a good cause.
> >> 
> >> A desktop or laptop will typically draw far less power than a single 60W
> >> incandescent bulb. I bet you have quite a lot of those. Even if not, the
> >> CFLs you'll have to give you light at night still draw much much more
> >> than a computer.
> >> 
> >> If saving energy is your personal driver, then you should be looking at
> >> water heaters, central heaters, aircon and stove as the main culprits.
> >> Everything else, whilst measurable, is a small drop in the bucket and
> >> probably not worth worrying about.
> >> 
> >> Assuming of course that your computer is a desktop/laptop, and not a 42U
> >> cabinet jam packed full of Dell 2950s
> > 
> > Don't forget the clothes dryer to, if you have one.  Mine is electric
> > and it pulls as much as my water heater does.  I just don't use it as
> > much is all.
> 
> I forgot about that :-)
> 
> Add in almost all laundry appliances and kitchen power tools too...

Modern appliances with Green stickers on them (whatever they're called) are 
more efficient by design.  To some extent this is also true with PCs.  I still 
have an old Pentium 4 32bit running a couple of test environments and back up 
storage.  I can assure you that the room gets hot after it has been running 
for a couple of hours!  :-)
-- 
Regards,
Mick

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 473 bytes --]

Re: [gentoo-user] system uptime

[Dale]

Mick wrote:
> On Sunday 30 Aug 2015 18:05:13 Alan McKinnon wrote:
>> On 30/08/2015 19:00, Dale wrote:
>>>
>>> Don't forget the clothes dryer to, if you have one.  Mine is electric
>>> and it pulls as much as my water heater does.  I just don't use it as
>>> much is all.
>> I forgot about that :-)
>>
>> Add in almost all laundry appliances and kitchen power tools too...
> Modern appliances with Green stickers on them (whatever they're called) are 
> more efficient by design.  To some extent this is also true with PCs.  I still 
> have an old Pentium 4 32bit running a couple of test environments and back up 
> storage.  I can assure you that the room gets hot after it has been running 
> for a couple of hours!  :-)


True.  My old puter, AMD 2500+ with 3GBs of memory, pulled at least
double if not more than my current 4 core AMD with 16GBs of ram.  I'm
not sure this new one has anything "green" on it but it is less power
hungry.  My old also helped heat my old room.  It had to be pretty cold
outside for me to turn the heat on.

My dryer tho, it's about 25 years old.  I think it pulls around 4500
watts normally.  Considering I have retired the heating element at least
5 or 6 times, it may pull a little more than that now.  I might add, it
takes longer to dry clothes tho.  I fear the day I can't tie that
element back together.  I doubt I will ever find a element for that old
thing.  I may have to get some new line for my old fashioned clothes
dryer.  You know, two trees with a wire between them.  A tree limb broke
my old one. 

Dale

:-)  :-) 

Re: [gentoo-user] system uptime

[Peter Humphrey]

On Sunday 30 August 2015 18:26:49 Mick wrote:

> Modern appliances with Green stickers on them (whatever they're called) are
> more efficient by design.  To some extent this is also true with PCs.  I
> still have an old Pentium 4 32bit running a couple of test environments and
> back up storage.  I can assure you that the room gets hot after it has been
> running for a couple of hours!  :-)

The desktop machine I'm referring to (an Amari "workstation") dates from 2009. 
It has an i5 processor, 16GB RAM* and two 2GB SSDs as the main power sinks. It 
sits (runs) in a boxroom 6ft square and keeps it comfortably warm. I haven't 
noticed any change in ambient temp since the SSDs replaced spinners.

* Whoever named that Random Access had a strange understanding of English. The 
last thing I want from memory is random access! How much better it would have 
been to call it something like Direct Access. Oh well - much too late now.

-- 
Rgds
Peter

Re: [gentoo-user] system uptime

[Alan McKinnon]

On 31/08/2015 10:50, Peter Humphrey wrote:
> On Sunday 30 August 2015 18:26:49 Mick wrote:
> 
>> Modern appliances with Green stickers on them (whatever they're called) are
>> more efficient by design.  To some extent this is also true with PCs.  I
>> still have an old Pentium 4 32bit running a couple of test environments and
>> back up storage.  I can assure you that the room gets hot after it has been
>> running for a couple of hours!  :-)
> 
> The desktop machine I'm referring to (an Amari "workstation") dates from 2009. 
> It has an i5 processor, 16GB RAM* and two 2GB SSDs as the main power sinks. It 
> sits (runs) in a boxroom 6ft square and keeps it comfortably warm. I haven't 
> noticed any change in ambient temp since the SSDs replaced spinners.
> 
> * Whoever named that Random Access had a strange understanding of English. The 
> last thing I want from memory is random access! How much better it would have 
> been to call it something like Direct Access. Oh well - much too late now.
> 


It's random access to distinguish it from serial access. In the early
early days there were a lot of strange methods being tried to build
memory - like dots on a cathode ray tube! To get to bit you wanted, you
had to wait till the scanning beam reached that part of the screen -
serial access. Addressable memory on a grid pattern came much later.

Random Access really means "able to access any random address as fast as
any other random address".

RAM is also not the opposite of ROM :-)

-- 
Alan McKinnon
alan.mckinnon@gmail.com

[OT] Was re: [gentoo-user] system uptime

[Peter Humphrey]

On Monday 31 August 2015 11:42:28 Alan McKinnon wrote:
> On 31/08/2015 10:50, Peter Humphrey wrote:
> > The desktop machine I'm referring to (an Amari "workstation") dates from
> > 2009. It has an i5 processor, 16GB RAM* and two 2GB SSDs as the main
> > power sinks. It sits (runs) in a boxroom 6ft square and keeps it
> > comfortably warm. I haven't noticed any change in ambient temp since the
> > SSDs replaced spinners.
> > 
> > * Whoever named that Random Access had a strange understanding of English.
> > The last thing I want from memory is random access! How much better it
> > would have been to call it something like Direct Access. Oh well - much
> > too late now.
> 
> It's random access to distinguish it from serial access. In the early
> early days there were a lot of strange methods being tried to build
> memory - like dots on a cathode ray tube! To get to bit you wanted, you
> had to wait till the scanning beam reached that part of the screen -
> serial access. Addressable memory on a grid pattern came much later.

Yes, of course I know all that, but it's still the antithesis of random - it's 
absolutely specific. Random is what you'd get if you didn't specify anything.

My favourite storage medium was core store. Millions of tiny ferrite rings, 
each at an intersection of orthogonal X and Y wires to specify the address, 
and a write pulse on another wire on the Z axis. At least, that's as close as 
I can remember now, 40 years later. No wonder computers were expensive.

I won't tell you what systems used a 24-bit processor and 12 or 16 KB of 2us 
core store backed by a 2MB disk (three feet in diameter), for fear of 
frightening you.    ;-)

> Random Access really means "able to access any random address as fast as
> any other random address".

My point is simply that the addresses are very far from randomly chosen. The 
distinguishing feature of the store is that you can go directly to the 
required location, without having to wait for it to reach the read/write 
device.

As I said though, there'd be no point in getting all stressed about it now.

> RAM is also not the opposite of ROM :-)

I seem to be having a senior moment here; at least, I don't follow that.

-- 
Rgds
Peter

Re: [OT] Was re: [gentoo-user] system uptime

[Alan McKinnon]

On 31/08/2015 15:41, Peter Humphrey wrote:
> On Monday 31 August 2015 11:42:28 Alan McKinnon wrote:
>> On 31/08/2015 10:50, Peter Humphrey wrote:
>>> The desktop machine I'm referring to (an Amari "workstation") dates from
>>> 2009. It has an i5 processor, 16GB RAM* and two 2GB SSDs as the main
>>> power sinks. It sits (runs) in a boxroom 6ft square and keeps it
>>> comfortably warm. I haven't noticed any change in ambient temp since the
>>> SSDs replaced spinners.
>>>
>>> * Whoever named that Random Access had a strange understanding of English.
>>> The last thing I want from memory is random access! How much better it
>>> would have been to call it something like Direct Access. Oh well - much
>>> too late now.
>>
>> It's random access to distinguish it from serial access. In the early
>> early days there were a lot of strange methods being tried to build
>> memory - like dots on a cathode ray tube! To get to bit you wanted, you
>> had to wait till the scanning beam reached that part of the screen -
>> serial access. Addressable memory on a grid pattern came much later.
> 
> Yes, of course I know all that, but it's still the antithesis of random - it's 
> absolutely specific. Random is what you'd get if you didn't specify anything.


Ah, an old timer - I forgot that for a second there :-)


> My favourite storage medium was core store. Millions of tiny ferrite rings, 
> each at an intersection of orthogonal X and Y wires to specify the address, 
> and a write pulse on another wire on the Z axis. At least, that's as close as 
> I can remember now, 40 years later. No wonder computers were expensive.

40 years maybe, but still dead on the money. That's exactly how that
memory worked.

> 
> I won't tell you what systems used a 24-bit processor and 12 or 16 KB of 2us 
> core store backed by a 2MB disk (three feet in diameter), for fear of 
> frightening you.    ;-)

Nah, I have some experience with such things.

Remember the old horror stories about not smoking in the computer room,
because smoke particles are much bigger than fly height of the disk
heads? The young 'uns here never had to deal with that.

> 
>> Random Access really means "able to access any random address as fast as
>> any other random address".
> 
> My point is simply that the addresses are very far from randomly chosen. The 
> distinguishing feature of the store is that you can go directly to the 
> required location, without having to wait for it to reach the read/write 
> device.

We understand each other perfectly; the odd bit is that word "random".
We both know it doesn't have the obvious meaning to a modern eye, and we
both know what random access really means

> 
> As I said though, there'd be no point in getting all stressed about it now.
> 
>> RAM is also not the opposite of ROM :-)
> 
> I seem to be having a senior moment here; at least, I don't follow that.

When I was still a kid learning about memory, many folks thought ROM was
very different from RAM, and that somehow ROM didn't have the same
random access qualities that RAM has. It does, except that ROM can't be
written (and dynamic RAM needs continual refreshing which ROM doesn't,
but that's another topic).

Eventually I gave up trying to clarify that part, but sometimes (like
now) the old habit comes back



-- 
Alan McKinnon
alan.mckinnon@gmail.com

[gentoo-user] Re: [OT] Was re: system uptime

[James]

Alan McKinnon <alan.mckinnon <at> gmail.com> writes:

> Ah, an old timer - I forgot that for a second there 

Both you guys are 'young whipper_snappers.....'


> Eventually I gave up trying to clarify that part, but sometimes (like
> now) the old habit comes back

Both of you have 'fuzzified' the key terms::
"Random vs Sequential" [1,2]. 

In SSD memory it's a hybrid, referred to as block access; so the issue is
still with us, just morphed.

hth,
James


[1]
http://kb.sandisk.com/app/answers/detail/a_id/8150/~/difference-between-sequential-and-random-access-operations


[2] https://en.wikipedia.org/wiki/Sequential_access_memory

Re: [OT] Was re: [gentoo-user] system uptime

[Peter Humphrey]

On Monday 31 August 2015 16:39:26 Alan McKinnon wrote:
> On 31/08/2015 15:41, Peter Humphrey wrote:

> > I won't tell you what systems used a 24-bit processor and 12 or 16 KB of
> > 2us core store backed by a 2MB disk (three feet in diameter), for fear of
> > frightening you.    ;-)
> 
> Nah, I have some experience with such things.

I thought I'd be misunderstood. I meant the purposes the computer systems were 
put to. In one case it was the closed-loop control of a nuclear power reactor 
(AGR); in the other the analysis and control of the national power grid. And 
the lights stayed on!

> Remember the old horror stories about not smoking in the computer room,
> because smoke particles are much bigger than fly height of the disk
> heads? The young 'uns here never had to deal with that.

Yes, of course. Them was the days - when we was young. Miles of 8-hole tape 
flying around the room. Entering boot code manually on key-switches. Two-day 
course in maintenance of ASR-33 and KSR-35. Back injury from manhandling a 
power supply into position. Aye...hmm...

> We understand each other perfectly;

Well, that must be a novelty :-)

-- 
Rgds
Peter

Re: [OT] Was re: [gentoo-user] system uptime

[Alan McKinnon]

On 31/08/2015 15:41, Peter Humphrey wrote:
> On Monday 31 August 2015 11:42:28 Alan McKinnon wrote:
>> On 31/08/2015 10:50, Peter Humphrey wrote:
>>> The desktop machine I'm referring to (an Amari "workstation") dates from
>>> 2009. It has an i5 processor, 16GB RAM* and two 2GB SSDs as the main
>>> power sinks. It sits (runs) in a boxroom 6ft square and keeps it
>>> comfortably warm. I haven't noticed any change in ambient temp since the
>>> SSDs replaced spinners.
>>>
>>> * Whoever named that Random Access had a strange understanding of English.
>>> The last thing I want from memory is random access! How much better it
>>> would have been to call it something like Direct Access. Oh well - much
>>> too late now.
>>
>> It's random access to distinguish it from serial access. In the early
>> early days there were a lot of strange methods being tried to build
>> memory - like dots on a cathode ray tube! To get to bit you wanted, you
>> had to wait till the scanning beam reached that part of the screen -
>> serial access. Addressable memory on a grid pattern came much later.
> 
> Yes, of course I know all that, but it's still the antithesis of random - it's 
> absolutely specific. Random is what you'd get if you didn't specify anything.


Ah, an old timer - I forgot that for a second there :-)


> My favourite storage medium was core store. Millions of tiny ferrite rings, 
> each at an intersection of orthogonal X and Y wires to specify the address, 
> and a write pulse on another wire on the Z axis. At least, that's as close as 
> I can remember now, 40 years later. No wonder computers were expensive.

40 years maybe, but still dead on the money. That's exactly how that
memory worked.

> 
> I won't tell you what systems used a 24-bit processor and 12 or 16 KB of 2us 
> core store backed by a 2MB disk (three feet in diameter), for fear of 
> frightening you.    ;-)

Nah, I have some experience with such things.

Remember the old horror stories about not smoking in the computer room,
because smoke particles are much bigger than fly height of the disk
heads? The young 'uns here never had to deal with that.

> 
>> Random Access really means "able to access any random address as fast as
>> any other random address".
> 
> My point is simply that the addresses are very far from randomly chosen. The 
> distinguishing feature of the store is that you can go directly to the 
> required location, without having to wait for it to reach the read/write 
> device.

We understand each other perfectly; the odd bit is that word "random".
We both know it doesn't have the obvious meaning to a modern eye, and we
both know what random access really means

> 
> As I said though, there'd be no point in getting all stressed about it now.
> 
>> RAM is also not the opposite of ROM :-)
> 
> I seem to be having a senior moment here; at least, I don't follow that.

When I was still a kid learning about memory, many folks thought ROM was
very different from RAM, and that somehow ROM didn't have the same
random access qualities that RAM has. It does, except that ROM can't be
written (and some RAM needs continual refreshing which ROM doesn't, but
that's another topic).

Eventually I gave up trying to clarify that part.



-- 
Alan McKinnon
alan.mckinnon@gmail.com

Re: [gentoo-user] system uptime

[Rich Freeman]

On Sun, Aug 30, 2015 at 11:56 AM, Peter Humphrey <peter@prh.myzen.co.uk> wrote:
> No-one has yet mentioned taking backups. I'm still using a brute-force
> approach, in which I shut down each of my two machines once a week to make a
> backup to external disk.

I have a multi-tier strategy.  Anything I'm going to complain about
losing is backed up daily to S3 with duplicity, end of story.  Stuff
like MythTV recordings and such which would be an inconvenience to
lose gets backed up daily to ext4 just in case btrfs gives out on me.
Otherwise I trust mirroring enough for that sort of thing.

If it isn't automatic, daily, and offsite, it isn't backed up as far
as I'm concerned.

-- 
Rich

Re: [gentoo-user] system uptime

[Michel Catudal]

Le 2015-08-30 11:56, Peter Humphrey a écrit :
> On Sunday 30 August 2015 00:04:43 Philip Webb wrote:
>> How long do desktop users typically leave their systems between reboots ?
>> How long between power off/on's ?
>>
>> I've long been in the habit of switching everything off while I sleep,
>> then restarting after I've woken & got going again myself.
>> However recently, I've run into delays getting my router
>> (only  1  device attached) to shake hands successfully with my ISP's server,
>> which have been requiring several power off/on's before it works.
>> As a result, I've started rebooting only after my weekly system update
>> -- it means I get to use the new versions of everything --
>> & not powering off at all ; the monitor + Xscreensaver are off
>> whenever I'm away from the machine for  >= 1 hr  (approx).
>>
>> Are there any pro's/con's I sb aware of ?
> No-one has yet mentioned taking backups. I'm still using a brute-force
> approach, in which I shut down each of my two machines once a week to make a
> backup to external disk. Otherwise they're on 24 hours a day running BOINC
> projects. On the desktop PC kmail makes a daily archive of messages, and once
> a day a cron job copies my user directory to /home/<me>.bu/ .
>
> I know it burns energy but I'm prepared to make my small contribution to what
> I think is a good cause.
>
Backups are vital for a server in company. At work we do a backup every day. At home, it depends how important your stuff is. For pictures you should always copy them on DVD. I regularly backup pictures for people who have ususable windows systems, for 
them the pictures are the most important stuff but they do not back them up.

Personally I don't like to do regular backups because that involves too many DVDs. I probably should do my backups more often.
I do have 3 2TB hard disks with important data copied on each for redudancy. I also have some backups on a 500G driver which is not powered usually. I also make some backup on DVDs sometimes.
Anything that is of extreme importance I have in several DVDs which I make copies of every few months. I remembered that in the early days of CD that their life was rather limited and am not taking chances on DVD even though I think the technology is a lot 
better.

-- 
For Linux Software visit
http://home.comcast.net/~mcatudal
http://sourceforge.net/projects/suzielinux/

Re: [gentoo-user] system uptime

[Terry Z.]

[-- Attachment #1: Type: text/plain, Size: 3263 bytes --]

I personally find it beneficial to backup to an online source locally or in
an online storage service (as long as encryption incurs etc).

DVD are indeed limited in life.  You are still better off with other
offline storage mediums such as an external hdd or tape indeed.

I've found crashplans unlimited storage 10 machine online backup solution
to be an excellent solution for desktop machines where connectivity is not
guaranteed for cronnd rsyncs etc. Of course it relies on running a fat jar
, but it works.

As to uptime, I keep my windows desktops machine online more than my linux
desktops just due to how frequent kernel updates occur.
On Aug 30, 2015 7:11 PM, "Michel Catudal" <mcatudal@comcast.net> wrote:

> Le 2015-08-30 11:56, Peter Humphrey a écrit :
>
>> On Sunday 30 August 2015 00:04:43 Philip Webb wrote:
>>
>>> How long do desktop users typically leave their systems between reboots ?
>>> How long between power off/on's ?
>>>
>>> I've long been in the habit of switching everything off while I sleep,
>>> then restarting after I've woken & got going again myself.
>>> However recently, I've run into delays getting my router
>>> (only  1  device attached) to shake hands successfully with my ISP's
>>> server,
>>> which have been requiring several power off/on's before it works.
>>> As a result, I've started rebooting only after my weekly system update
>>> -- it means I get to use the new versions of everything --
>>> & not powering off at all ; the monitor + Xscreensaver are off
>>> whenever I'm away from the machine for  >= 1 hr  (approx).
>>>
>>> Are there any pro's/con's I sb aware of ?
>>>
>> No-one has yet mentioned taking backups. I'm still using a brute-force
>> approach, in which I shut down each of my two machines once a week to
>> make a
>> backup to external disk. Otherwise they're on 24 hours a day running BOINC
>> projects. On the desktop PC kmail makes a daily archive of messages, and
>> once
>> a day a cron job copies my user directory to /home/<me>.bu/ .
>>
>> I know it burns energy but I'm prepared to make my small contribution to
>> what
>> I think is a good cause.
>>
>> Backups are vital for a server in company. At work we do a backup every
> day. At home, it depends how important your stuff is. For pictures you
> should always copy them on DVD. I regularly backup pictures for people who
> have ususable windows systems, for them the pictures are the most important
> stuff but they do not back them up.
>
> Personally I don't like to do regular backups because that involves too
> many DVDs. I probably should do my backups more often.
> I do have 3 2TB hard disks with important data copied on each for
> redudancy. I also have some backups on a 500G driver which is not powered
> usually. I also make some backup on DVDs sometimes.
> Anything that is of extreme importance I have in several DVDs which I make
> copies of every few months. I remembered that in the early days of CD that
> their life was rather limited and am not taking chances on DVD even though
> I think the technology is a lot better.
>
> --
> For Linux Software visit
> http://home.comcast.net/~mcatudal
> http://sourceforge.net/projects/suzielinux/
>
>
>

[-- Attachment #2: Type: text/html, Size: 4021 bytes --]

Re: [gentoo-user] system uptime

[Philip Webb]

Thanks for the responses to my inquiry re how long systems stay up.

My feeling for electronics is that equipment needs stability :
temperature + humidity should not vary much
& electrons need to flow thro' wires regularly.
That suggests a machine will do better if left running.

I don't like the idea of phoning my ISP's help desk
with an ill-defined problem : I can't be sure it will recur
if they ask me to go thro' a few restart cycles ;
the staff are probably trained to offer a few simple solutions ;
they may simply offer to send me a new router
or worse to send me a modem instead, which is harder to set up.
The problem probably lies in their server :
my router enjoys constant conditions & shows no other signs of wear ;
if that's the case, ISP help staff may not even know what has changed.

I'm planning to build a new machine in the next few weeks,
so I also now plan to set it up to do suspend, hibernate etc ;
I've grabbed a few dox which explain what's involved.
Mick mentioned 'S3' = 'sleep' & 'S4' = 'hibernation' in this connection :
what do they mean ?  Do they correspond to saving to RAM/disk ?

-- 
========================,,============================================
SUPPORT     ___________//___,   Philip Webb
ELECTRIC   /] [] [] [] [] []|   Cities Centre, University of Toronto
TRANSIT    `-O----------O---'   purslowatchassdotutorontodotca

Re: [gentoo-user] system uptime

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 645 bytes --]

On Thu, 3 Sep 2015 11:32:52 -0400, Philip Webb wrote:

> I'm planning to build a new machine in the next few weeks,
> so I also now plan to set it up to do suspend, hibernate etc ;
> I've grabbed a few dox which explain what's involved.
> Mick mentioned 'S3' = 'sleep' & 'S4' = 'hibernation' in this
> connection : what do they mean ?  Do they correspond to saving to
> RAM/disk ?

Short answer: Yes

Long ansder:
https://en.wikipedia.org/wiki/Advanced_Configuration_and_Power_Interface#Power_states


-- 
Neil Bothwick

** I'm not going to get married again **
** I'll just find a woman I don't like and give her a house **

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

Re: [gentoo-user] system uptime

[Philip Webb]

150903 Neil Bothwick wrote:
> On Thu, 3 Sep 2015 11:32:52 -0400, Philip Webb wrote:
>> Mick mentioned 'S3' = 'sleep' & 'S4' = 'hibernation' in this connection:
>> what do they mean ?  Do they correspond to saving to RAM/disk ?
> Short answer: Yes
> Long answer:
> https://en.wikipedia.org/wiki/Advanced_Configuration_and_Power_Interface#Power_states

Thanks.

-- 
========================,,============================================
SUPPORT     ___________//___,   Philip Webb
ELECTRIC   /] [] [] [] [] []|   Cities Centre, University of Toronto
TRANSIT    `-O----------O---'   purslowatchassdotutorontodotca